JPS63153045A - Method and apparatus for detecting change of system activity - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to the determination of activity changes that occur within biological systems in response to applied stimuli, or to tasks attempted in connection with this. In particular, the present invention relates to a method and apparatus for detecting changes in brain activity displayed in electrical brain wave signal waveforms in response to stimulation applied to the brain.

(Prior Art) In electroencephalogram (recording) testing methods, minute electrical signals generated in the brain are monitored, analyzed, and often recorded.

The interpretation of such signals is the basis of neurological research and clinical diagnosis of neurological diseases.

The electroencephalograph measures the potential on the surface of the subject's scalp, and in this case, electrodes are bonded to the scalp surface at one or more locations. This adhesion is called the "10/20 system" and is based on the International Union of Electroencephalography (International Union of Electroencephalography)
Federation of Elcctrocnc
cpharography).

Typically, when used for diagnosis, there may be as many as 20 electrodes arranged and bonded in this manner.
Each electrode is connected to an electroencephalogram device and the electrical potential is measured and instructed. These potentials typically range from 1 to 100 μV and are spontaneous but often controlled by an electroencephalography device (e.g., are not perceived by the subject's eyes). measured in relation to the occurrence of transitional light patterns that should occur.

These electroencephalogram (EEG) signals, ie, the measured electric potentials, have different frequency content (components) depending on the activity in biological systems including the brain. This frequency content is divided into four basic frequency bands: (1) the "delta (δ)" band from 0-4 Hz;
2) 4-8Hz “theta (θ)” band, (3) 8-1
3Hz “alpha” band and (4) 13Hz
It is classified into the above “beta (β)” band.

A typical type of information that is desirable to obtain from an EEG signal during a specific time period is the dominant frequency contained in a particular signal during the same time period;
It requires considerable training and is highly dependent on the proficiency of the neurologist.

The reason is that an electroencephalogram signal usually contains many frequency components.

This analysis can be made more convenient and modified by the use of signal processing equipment to obtain parameters and characteristics of the data obtained in such electroencephalogram signals.

For example, co-feeding one or more electroencephalogram signals to a suitably programmed computer can provide an analysis of the frequency vectors contained in such signal(s).

In the above-mentioned electroencephalogram measuring device, the electroencephalogram signal, which is an analog signal, is sampled over a selected time interval, and each obtained sample is converted into a digital value. and stored at least temporarily on the computer.

The digitized consecutive samples consist of a fast Fourier transform (F F
T') algorithm.

As a result of this transformation, a frequency spectrum representing the frequency content is displayed in the measured signal sample. When such a spectrum is expressed using frequency as an axis, it can be displayed as a graph with amplitudes of frequency components included in the signal.

A discontinuous Fourier transform is defined for such digitized samples, which may use the FFT algorithm, but
The reason for this is that in the case of the above-mentioned (electroencephalogram measurement) device, the corresponding frequency spectrum can be obtained relatively quickly by reducing the required amount of calculations.

These samples could be collected to obtain the frequency spectrum and conventional Fourier methods could be used, but this would increase the complexity of the equipment and the time required. Nevertheless, the latter device has the advantage that the time interval over which the signal is measured and sampled does not influence the results.

However, using the former device to speed up the transition to the frequency domain introduces a limit on the minimum time that data must be acquired. This is because the length of the interval determines the period of the lowest frequency in the frequency spectrum resulting from the use of the Fourier methods described above.

EEG signals contain very low frequencies that are of great interest to neurologists, so the interval between data acquisition must be long enough to obtain these frequencies. .

If a shorter time interval is used to obtain the data,
The lowest frequency that can be included in the spectrum will be a larger frequency value and therefore information of interest will be lost.

For example, a 1 second time interval may present frequencies as low as about IHz, whereas a 20 millisecond data collection will present a lowest frequency of about 50 Hz.

Activity changes in a biological system in response to a stimulus occur over a duration significantly less than 1 second, but the frequency content within the signal representing this activity change must include frequencies of interest within the IHz range. Therefore, the above-mentioned frequency domain representation becomes an important issue.

Thus, while it is desirable for a device to show changes in system activity that occur within a reasonable time, such as a fraction of a second, after stimulation, it is important to note that the These are changes that include frequency components with a period longer than a given duration and are included in a signal that represents information about these changes.

SUMMARY OF THE INVENTION It is an object of the invention to provide a means for taking samples of signals indicative of activity in a system, whether spontaneous or in response to a stimulus, and for receiving these samples and Signal processing f9 to capture relative magnitude relationships with nearby specimens as a basis for displaying the occurrence of changes in system activity signals.
The goal is to provide the following.

The number of these changes that occur over successive fractions of the data acquisition period provides an indication of the amount of change that occurs at each point in time over the data acquisition interval. If there are repeated system signals that are useful for similar situations, these activity changes at each point in time are
It can be averaged and used as a more reliable indication of activity at each point in time.

The results thus obtained can be useful to the user in any display device that has been found useful for the above purpose.

(Embodiments of the Invention) FIG. 1 is a view of a subject's head 10 viewed from above, and shows the standard arrangement of electrodes for electroencephalogram testing, which are placed on the scalp surface. Each electrode position is designated by commonly used symbols: Fpl, Fp2°FQ, F3 . F4. F7. F
B,,T3. T4゜T5. T6. CO, C3, C4, P
They are distinguished by O, P3°P4.01 and 02. Also shown are locations A1 and A2 for attaching both reference electrodes, which would normally be attached to one or both of the subject's ears.

In FIG. 1, an electroencephalogram signal analysis system 11 is placed on the right side of the head 10. The analysis system 11 has an electroencephalogram electrode array module 12 from which extends a group of connecting cables 13, which are shown suitably short and simplified.

Typically, 16 signal-introducing electrodes are located at the far end of each connecting cable 13. A further reference electrode is present at one end of the cable 13 and is usually glued to the ear, but can also be connected to other parts and is often used.

These electrodes are of commercially available standard specifications commonly used in electroencephalogram examinations, and are therefore omitted from illustration. Each electrode not shown should be glued to the head 10 in one of the positions shown. Depending on the content of the inspection,
The number of electrodes may be small or large depending on the needs of the data collected for the test.

The signals obtained by the electrode array 12 are transmitted to an amplification system 14, which is equipped with one amplifier for each cable 13. Each amplifier is typically a differential amplifier and has a corresponding connection cable 13
The signals transmitted via the head 10 at positions A1 and A
The electrodes attached to one or both of 2 are used as a reference level.

These amplifiers provide gains in the order of 70,000 and amplify signals containing frequencies below tens of Hz. In this case, no signal degradation occurs due to any response limitations of the amplifier. These amplifiers are well known for use in electroencephalogram testing, so their explanation will be omitted.

The amplified analog brain wave signal is sent to an amplification module 14.
to an analog multiplexer and then to an analog-to-digital converter contained within conversion module 15. Here, successive samples over selected time intervals of the amplitude of the signal captured by each electrode are obtained and their corresponding digital values are obtained in a conversion module 15 in a known manner.

If the analog sample is to be converted to a digital representation of a pattern consisting of 14 magnitude bits and 1 sign bit, the conversion module 15 should provide a resolution reasonable given the state of the art. is recognized.

As is well known, sampling is performed repeatedly at fixed time intervals or frequencies, and the sampling rate (frequency) is the highest frequency in the electroencephalogram signal that should be represented by these samples. Must be more than double.

Therefore, if the upper frequency content is unclear, the sampling rate should be increased to the point where there is no information about the signal beyond twice the highest rate. From this perspective, a typical sampling rate for module 15 is 2
56 Hz/sec, which is reasonable given the current state of the art.

As a converter having the performance described here, a well-known and commonly commercially available converter is sufficient, and further detailed description thereof will be omitted here.

The digitized samples processed in the conversion module 15 are sent to the signal processing means 16. The respective digitized samples of the signals obtained from each position on the scalp 10 are analyzed there, and under the control of this signal processing means 16,
Changes in brain activity that occur in the head 10 in response to stimulation applied to the head 10 from the stimulation source 17 are determined.

A portion of a typical electroencephalogram waveform transmitted through the data collection cable of the cable 13 is shown in FIG. In this figure, the origin along the time axis indicates the time point of stimulus application. The measured potential for the signal is expressed as the voltage “V” on the vertical axis.
is plotted as. A thin perpendicular line that intersects the waveform is drawn on the time axis, and the sample number corresponding to the waveform position is written on the extension of the line.

The time and vertical axes are interrupted near their respective origins, meaning that only a particular portion of the example waveform was chosen for illustration. That is, the selected sample numbers 105 to 121 are merely selected as typical examples and have no further meaning.

The entire waveform typically spans from a few tenths of a second to two seconds. The measured digital values for sample numbers 105 to 121 are not shown in FIG. 2 because they are unnecessary for the purpose of explanation.

Changes in brain activity are displayed by electroencephalogram signal waveforms,
This waveform tends to have the characteristics of a low-frequency content and broadband wave, with many higher-frequency content overlapping waves occurring along it, and thus changes in brain activity essentially occur over the waveform interval. is indicated by an increase or decrease in the number of overlapping wave peaks in different time segments.

That is, the increase or decrease in the number of local minima and maxima in a short time window along the time axis determines the activity change in question, i.e., the activity change in the brain's biological system that occurs spontaneously or in response to stimulation. Is displayed. Such activity changes are related to information (
cortical processes of the cerebral cortex
It is thought that it is related to ``sing''.

Therefore, it is of great importance to measure the number of maxima and minima that occur in short time segments for each point along the time axis.

Changes in the number of local maxima and minima occur at frequencies significantly lower than the frequency content of the overlapping wave peaks themselves. However, as mentioned above, it is not possible to use a frequency domain representation for very short periods of time as an indication of the number of changes that have occurred in the signal within the analogous window along the axis. The reason is that the lower limit of frequencies to be displayed by use of the FFT algorithm will exceed the frequency content of the rate of change of the number of overlapping waveform peaks.

Thus, for example, if we check for activity content in a 20 ms window along the time axis, whatever is measured in this window, the FFT
When an algorithm is processed and the results are displayed, no meaningful information is obtained.

Figure 2 illustrates a very elegant alternative method that both precisely finds the slope characteristics of the waveform component for each sample point and also It is based on counting inflection points with different slopes on either side.

This means that, on the basis of the digitized samples in the signal processing module 16, each digitized sample is positioned immediately adjacent on either side of the sequence of successive digitized amplitude samples obtained from the conversion module 15. This can be accomplished by considering the digitized specimen.

Each sample must be placed on either side of it, i.e. before it or Its amplitude value will be compared with the digitized sample generated later. Each such digitized sample has a relative magnitude relationship with the two digitized samples located adjacent to each other, i.e., is greater than, smaller than, or equal to the value of the adjacent digitized sample. There will be a relationship like this.

If the size of the center sample among these three digitized samples is larger than the sample at the previous position, the waveform between them is interpreted to have a positive slope. A "10" symbol is written below the number row to indicate the slope.

Also, if the central digitized sample has a smaller value than the preceding sample, its slope is negative; if they have the same value, the slope is zero.

Similarly, if the size of the central digitized sample is larger than that of the later adjacent digitized samples, the waveform between them is said to have a negative slope, and if it is smaller, the slope is positive. . Of course, if their magnitudes are the same, the waveform will have zero slope.

A positive slope is a “10” sign, a negative slope is a “-” sign,
By indicating "O" in the case of no tilt, the results for each successive pair of digitized samples in the digitized sample sequence obtained over the acquisition interval during which the electroencephalogram signal was measured can be confirmed. This code string is shown in the "slope instruction" row of FIG. 2, which shows an example waveform.

A local minimum or maximum, or inflection point, is believed to occur in FIG. 2 wherever the slope sign changes as one progresses along the slope indication line. In other words, the point where such a change occurs in the relative size relationship between the central digitized specimen and the two specimens immediately before and after it is an indication of an inflection point.

Furthermore, the judgment number is 1, and the number "1" is assigned to each change in the subsequent code that occurs along this slope instruction row, while "1" is assigned when no change occurs in the code string of this row.
.theta.'' is given.The sequence of judgment numbers obtained in this manner is shown in FIG. 2 as a judgment number string.

By creating this judgment number string, the number of local minimum and local maximum changes can be measured for each point along the time axis in FIG. This count defines the time window for that point and
It is based on determining the number of local minima and local maxima that occur within that window.

This window is then shifted in the same relationship with respect to the next sample point, and local minima and maxima are counted in that window as an indication of the activity occurring at the subsequent sample point.

This is the substring (s
ubscqucncc), it can also be done based on the judgment number string. This subcolumn is often
The activity instructions will be chosen to be concentrated near the relevant judgment numbers at the data points to be tracked,
That doesn't necessarily have to be the case.

In FIG. 2, the first window subcolumn is arranged such that there are three subcolumn entries to the left and four to the right of the decision digit corresponding to the digitized specimen for which an indication of activity change is to be given. selected. The reason is that a window consisting of eight decision numbers was chosen as the window size in which the activation instruction should be given.

The span of the time window chosen is: I) too large and therefore does not average or practically provide sufficient resolution of changes in activity at each point along the time axis; 11) Because it is small to J degree,
It is a compromise between too many undefined changes in the results due to other uncontrollable events occurring in the system. A typical window size in electroencephalography would be 20-50 milliseconds, as shown schematically in FIG.

Note that the actual value was 31 milliseconds at a sampling rate of 256 Hz/second.

The first window sub-column is extracted from the judgment digit string and repeated in a separate row in FIG. This first window subcolumn contains the activity measures to be presented for sample point 109.
re).

The window sub-columns are connected by vertical lines to show their relationship to their corresponding test digit string portions.

The second window subcolumn for providing the activity measure for sample point 110 is shown in FIG. 2 as the second window subcolumn row. In order to establish a relationship between each component number of this second window sub-column and the corresponding determination number in the determination value string that displays them, they are connected by a thin corner-bent line.

This second window sub-sequence is the first one in the judgment number string.
Compared to the window sub-column, it starts with the judgment digit that is one lag behind and ends with the digit that is one lag behind.

Similarly, each window column for each sample point in the sample column is formed sequentially from the corresponding judgment digit string, but in FIG. 2 only the first and second window sub-columns are illustrated. .

Within a window as a subsequence selected along the time axis, a measure of system activity change should be considered in determining the system activity change that occurs in the vicinity of the sample point associated with the window; This window provides a measure of system activity change formed from the decision numbers contained therein.

This sub-sequence was formed as an arithmetic combination of the decision numbers therein to provide a measure of activity change. Simply, this can be the sum of the decision numbers present in a sub-column, such that the count of local maxima and minima occurring within the span with respect to the corresponding decision number and sample number is Obtained based on the judgment number.

Using this measure of activity change, the first window sub-column gives the value "4", which is listed as the first entry of the index number sequence at the bottom of FIG. This index number is written directly below its corresponding judgment number and sample number "109".

From the second window sub-column, using the scale described above, we obtain the index number "3", which corresponds to its corresponding judgment number and sample number "1" in the row of the index number sequence in Figure 2.
It is written directly below “10”. In the row of this index sequence, many entries are successively written,
These are for the window sub-sequences that should follow each of the first and second window sub-sequences, and are obtained by similar counting from successive numbers in the judgment number sequence.

This index number sequence then provides a measure of system activity changes that occur within the window for the sample to which the index number corresponds. In this way, after the subject's brain is stimulated by the index number sequence/time display,
A measure of system activity changes that occurred during the time the data sample was taken is provided. This display, which may be projected onto a video monitor or a simple recorder, provides the user of the analyzer system with a clear measure (indication) of changes in system activity in response to stimuli applied to the biological system. However, in this case, the user has to go through a great deal of interpretation (1n) to obtain such information.
There is no need to perform a terpretat 1on).

This index number sequence can also be obtained in other ways if required by the user. It is based on the judgment that different arithmetic combinations of test numbers in sub-columns are more appropriate for the requirements of testing to quantify changes in system activity in response to stimuli. For example, weighting is a modulus (veigl+ting), where a different weight (emphasis) is given to the numbers that make up the judgment number string than to other numbers.
function) could be applied.

Thus, 1 is not uniform, but rather differs for some of its sub-rows from others, i.e.
According to a quadratic windowing method, windows can be formed within the sub-column windows that are given different weights depending on their position.

One of the typical examples is the fourth power law of cosine (a
cosine to the lower powe
Blackman (Blackωa) who follows r law)
n) would be a window function. This function gives a full weight (fu
ll weight ), but relatively little weight is given to the judgment numbers at both ends of the same sub-column.

Due to the complexity of the human brain, changes in activity in response to a stimulus have many other co-occurring contributors that are not necessarily related to the stimulus. Quite commonly, therefore, a stimulus response test for a subject will be repeated many times, perhaps as many as 25 to 100 times.

In such a test, a new series of index values may be generated each time at the same reference point in time, over the same interval taken for the same stimulus, and with the same sampling rate. As a result,
Each of the corresponding index values in the results of each test may be averaged by further arithmetic combinations.

This provides a final measure (indication) of system activity changes. This has large variability due to other factors that are canceled out through the averaging process, but in this case the activity changes originally caused by the stimulus are also enhanced by the same averaging process.

This result is then displayed for a single test index sequence as described above.

In the clinic, the typical purpose of tests such as those described above is to determine whether there are changes in the brain's response to stimuli caused by an injury or a disease such as Alzheimer's disease. .

Therefore, stimuli are chosen that are known to affect two parts of the brain. Control group (cont
Measurements performed on each of the groups ( rol group ) showed that the elapsed time between the activity change in one part and the other part relative to the stimulus was constant. On the other hand, in measurements on subjects suspected of having deteriorated brain function, the same stimulation caused
It was observed that there was a difference in elapsed time between the activity changes occurring in the two brain regions mentioned above.

Thus, unlike in the case of a healthy control group, brain deterioration can be determined by the presence of a significant elapsed time difference between the changes in activity in the two regions of the subject's brain.

The signal processing means (module) 16 may be a general-purpose computer used in a laboratory, or a microprocessor installed in an analysis equipment unit in a small-scale diagnostic laboratory. However, in some clinics, a digital system may be desirable as it is believed to perform the functions described above.

An example of the arrangement of the signal processing means 16, in particular of a special system for processing digitized samples obtained from a single signal, is shown in FIG.

The digital system of FIG. 3 may be implemented in a multiplexing scheme that may be repeated for each signal source, or for all of the signals, if multiple signals are sent from one module 12. It could be used to process by For clarity of illustration, timing and control circuits that would be obvious to those skilled in the art have been omitted.

From the conversion module 15, the digitized samples are sent to a second register 21, where each sample is stored for one sampling period. In the next sampling period, the digitized samples stored in the register 20 in the previous sampling period are sent to and stored in another storage (second) register 21, and the new digitized samples are stored in the register 21. 20 is stored.

Similarly, the digitized sample stored in the second register 21 is sent to a further storage (third) register 22 for storage. The contents of the third register 22 are either discarded or sent to further memory, if desired, as shown in FIG. At each sampling period, the respective digitized samples stored in both registers 20 and 21 are sent to a digital (first) comparator 23 . The respective digitized samples stored in both registers 21 and 22 are sent to a second comparator 24.

In the comparators 23, 24, the relative magnitude relationship previously described between the digitized samples stored in register 21 and the digitized samples stored in registers 20 and 21 is determined.

The results of these comparisons, ie, whether the digitized sample in register 21 is greater, less than, or equal to the digitized sample in the register on either side thereof, are sent to decision logic module 25. . Therefore, the determination logic module 25 determines whether the above-mentioned magnitude relationship has changed.
It is determined whether it is "1" or "0" indicating whether there is a change or no change, and this result is output as the next entry in the judgment number string shown in FIG.

Both comparators 23, 24 thus make the relative magnitude decisions necessary to actually determine whether a slope representation, positive, negative or zero, should be made between successive adjacent digitized sample pairs. Enter module 25. The logic module 25 operates based on these relative magnitude determinations and determines a determination number string.

The output of register 25 is sent to a shift register 26 having as many shift positions as are required for the sub-columns that actually form a window along the time axis. Therefore, for the conditions described with respect to FIG. 2, shift register 26 will be provided with eight shift positions.

At each new sampling period, a new value is added to logic module 25.
is obtained from and transferred to the left end of shift register 25, while the final value at the right end is shifted out. This final value is either discarded if no longer needed, or sent to another suitable memory if the test digit string is to be retained, as shown in FIG.

Therefore, shift register 26 always has a sub-column for one window within it, such as the first window sub-column shown in FIG. In the next sampling period, a second window sub-column will appear in the shift register. Similarly, in subsequent sampling periods, i new window sub-columns will be formed in shift register 26.

Shift register 26 also includes a parallel output to multiplexer 27. Multiplexer 27 selects each storage location output from shift register 26 to be sent to multiplier 28 in succession. Multiplexer 27 is driven by a counter and logic means 29, which is reset at the beginning of each sampling period.

Multiplier 28 receives each test digit in the sub-column and multiplies it by a second window weighted number nine function stored in weight memory 30 and represented by the weight value.

These weight values may be calculated to provide a Blackman window as described above.

Of course, if a secondary window is not used, weight memory 30 and multiplier 28 may be omitted.

The weighted sub-column numbers are then sent to a sequential adder 31. A successive adder 31 receives all of the sub-column digits and sums them once for each sampling period, corresponding to that sampling period, in other words, corresponding to the samples taken within that sampling period. form an index sequence.

If the experiment is to be repeated, a further device section is provided for transferring the index sequence generated by the successive adder 31, namely an addition/storage unit consisting of an adder 32 and an index sequence storage memory 33. be done. Furthermore, the third
In the figure, as suggested by the arrow pointing to the right from the output of the successive adder 31, the index number sequence obtained from any one experiment can be stored separately.

The adder 32 receives each index number from the sequential adder 31 and retrieves from the memory 33 all the sums of the corresponding index numbers determined in previous experiments stored therein, combines them, and calculates the sum of the index numbers. The results are sent back to memory 33. A counter/logic means 34 (which is reset each sampling period) supplies the memory 33 with the proper address.

Another counter 35 keeps counting the number of experiments and its count is sent to a divider 36. The divider 36 divides the sum for each index number sequence position held in the memory 33 by the number of experiments at the end of the experiment based on repeated use of the sting to perform a series of experiments.

These results are then sent out of the divider 36 as the output of the module 16, as indicated by the arrow pointing to the right thereof. This output from module 16 is sent to display module 18 (see FIG. 1),
For example, it is transmitted to a video terminal.

Many other devices can be used in place of the one shown in FIG. 3 to accomplish the same purpose. Also, as mentioned above, the digital computer can be programmed to perform the operations to be performed by the third digital system. The programs necessary to cause such a computer to perform the steps necessary to generate a test number sequence, an index number sequence, and an average of the index number sequence can be easily accomplished by a skilled computer programmer.

Although the present invention has been described in terms of preferred embodiments, those skilled in the art will recognize that changes may be made in form and detail without departing from the spirit and scope of the invention.

[Brief explanation of the drawing]

FIG. 1 is a top view of a subject's head showing a typical arrangement of electrodes, and a block diagram showing a signal path of an electroencephalogram signal analysis device according to the present invention; FIG. 2 is a diagram showing a typical waveform of an electroencephalogram signal. A diagram illustrating one component, with the results of sampling and signal processing attached below; and FIG. 3 is a block diagram showing the block diagram of FIG. 1 in more detail. DESCRIPTION OF SYMBOLS 11... Brain wave (a system, 12... Electrode array module, 14... Amplification system, 15... Conversion module, 16... Signal processing means, 17... Stimulation source, 18... -Display module, 20...second register, 21...second register, 22...third
Register, 23... First comparator, 24... Second comparator, 25... Judgment logic module, 26... Shift register, 27... Multiplexer, 28... Multiplier 1.29 ... Counter/logic means, 30... Weight memory, 31... Sequential adder, 32... Adder,
33... Index number sequence storage memory, 34... Counter/
Logical means, 35... Counter, 36... Divider Patent attorney Michito Hiraki and 1 other person ■

Claims (20)

[Claims] (1) A system activity change detection method for determining an activity change occurring in a system, wherein the system activity change is indicated by at least one signal representative of such system activity, comprising: taking a first series of consecutive samples of amplitude values of the system activity signal sampled at a rate sufficient to represent a portion of the activity signal related to the activity change;
Each sample in the first column is taken for a selected duration set with respect to the selected time point; determining which of the first, second, and third relative size relationships occurs between the specimen and each specimen located adjacent to the specimen immediately before or after the specimen; forming a second column of numbers, each of these test numbers having a one-to-one correspondence with each sample in said first column except for both the first and last sample in said first column; Along with having,
If the corresponding sample in the first column has the same relative magnitude relationship with each of the samples located immediately before and after it, each judgment number is 1
However, if the corresponding sample in the first column has a different relative magnitude relationship with the samples located immediately before and after it, then the value of the selected numerical value set a second value; and generating an indication of the activity change based on the determination number. (2) The method according to claim 1, wherein the system activity signal is an electroencephalogram signal. (3) Following the formation of the second column, form a third column of index values, each index value corresponding to one of said decision numbers in said second column; The indicator value is
having a value equal to the selected first arithmetic combination of determination numbers constituting a sub-sequence of the selected determination numbers including the corresponding determination number, and further indicating the activity change based on the index value; 2. A method according to claim 1, characterized in that: (4) The selected sub-sequence of judgment numbers is characterized by being comprised of a group of continuous judgment numbers having approximately equal numbers of judgment numbers on the front and rear sides of the judgment number corresponding to the index value. The method according to claim 3. (5) A selected plurality of the determination values in said selected sub-column have the values determined in the formation of said second column, and said first column consisting of said determination values Before forming the selected combination, the order number is
4. A method according to claim 3, characterized in that a number) is formed. (6) The steps of taking the first sample row, determining the relative size relationship, forming the second row, and forming the third row constitute the initial steps, and is repeated with respect to said system at a point in time to form a first set of iterative stages, followed by a selected set of each of said index values in said initial stage with corresponding index values in said first set of iterative stages. A second arithmetic combination is formed, and the instruction for changing the activity is generated based on the selected second arithmetic combination.
The method described in section. (7) The method according to claim 3, wherein the system activity signal is an electroencephalogram signal. (8) A selected plurality of the determination numerical values in said selected sub-column have the numerical values determined in the formation of said second column, and said first column consisting of said determination numerical values An ordinal number is formed by subtraction before forming the selected combination, and a selected plurality of the judgment values located further away from the judgment value corresponding to the index value are further subtracted. The method according to claim 4. (9) The steps of taking the first sample row, determining the relative size relationship, forming the second row, and forming the third row constitute the initial steps, and is repeated with respect to said system at a point in time to form a first set of iterative stages, followed by a selected set of each of said index values in said initial stage with corresponding index values in said first set of iterative stages. 8. A second arithmetic combination is formed, further comprising: generating an instruction for changing the activity based on the selected second arithmetic combination.
The method described in section. (10) Detection for determining activity changes occurring in the subject system, such that the activity changes in the subject system are indicated by at least one signal representative of the activity of such subject system. In an apparatus, signal acquisition means for acquiring an activity signal of said system; analog-to-digital conversion means for providing a first series of successive digitized samples of amplitude values of said system activity signal; and a determination number. a second column consisting of digits, each of which has a one-to-one relationship with the sample in the column, excluding both the first and last sample in the first column. and each judgment value is selected if the corresponding sample in the first column has the same relative size relationship with each of the samples located immediately before and after it. However, if the corresponding sample in the first column has a different relative magnitude relationship with each of the samples located immediately before and after it, the selected numerical value An activity change detection device comprising: signal processing means for forming a second column to have a second value of the set; and display means for generating a display of the change in activity based on the determination number. (11) The device according to claim 10, wherein the system activity signal is an electroencephalogram signal. (12) The signal processing means may further form a third column consisting of each index value corresponding to the determination number in the second column, and each of the index values corresponds to the determination number corresponding to the determination number in the second column. having a value equal to the selected arithmetic combination of the test numbers constituting the selected test number sub-sequence, comprising a numeric value, and wherein said display means generates an indication of said activity change based on said indicator value. 11. The device according to claim 10, characterized in that: (13) The apparatus according to claim 10, wherein the signal processing means comprises computer means. (14) storage means for said signal processing means to at least temporarily store said digitized samples provided by said analog-to-digital converter means; and said digitized samples stored in said storage means. and for the selected digitized specimens so compared, the values of the selected other digitized specimens to be compared stored in the memory means. Is it larger or smaller compared to the value?
comparator means capable of generating an indication that the selected indication, which is generated by the comparator means, has a value of either equal or equal; logical means for generating an indication that is either the same as or different from the selected indication; and a logical means for at least temporarily storing the selected indication of the logical means. 11. The apparatus according to claim 10, characterized in that it comprises a storage means. (15) The signal processing means comprises a first, a second and a third, each of which is capable of storing the digitized sample.
registers, the first register successively receiving and temporarily storing each of the digitized samples sent by the analog-to-digital conversion means, and then registering each of the digitized samples currently received therein. The digitized sample located immediately before the digitized sample is transferred to the second register for temporary storage, and the second register continuously transfers the digitized sample transferred from the first register. after receiving and temporarily storing therein, transferring the digitized specimen immediately preceding each of said digitized specimens currently received to said third register for temporary storage therein; The three registers successively receive and temporarily store therein the digitized samples transferred from the second register, and the device further compares the values of the digitized samples each sent. a signal indicating whether one of the digitized samples has a value greater than, less than, or equal to the value of the other one sent to the comparator; comprising first and second comparator means capable of generating at its output, said first comparator means being connected to each of said first and second registers to present each register for comparison. said second comparator means is adapted to receive an indication of said digitized sample stored in said register, and said second comparator means is connected to said second and third registers to present said respective register for comparison. The signal processing means is configured to receive a representation of the stored digitized sample, and the signal processing means further determines whether the signals sent thereto are of the same sign or of different signs. ,
comprising logic means capable of generating a signal at its output indicating which condition is currently occurring, said logic means being electrically connected to said output of each of said first and second comparators; Said signal processing means receives a current signal component at its input for temporary storage and stores each successive current signal component along its storage location as each successive current signal component is received at said input. , wherein the input of the first shift register means is electrically connected to the output of the logic means, and each of the storage parts is connected to the other means. 11. The device according to claim 10, characterized in that it has: (16) The signal processing means comprises a first, a second and a third, each of which is capable of storing the digitized sample.
registers, wherein the first register successively receives and temporarily stores therein each of the digitized samples sent by the analog-to-digital conversion means, and the first register receives and temporarily stores therein each of the digitized samples currently received therein. The digitized sample located immediately before each digitized sample is transferred to said second register for temporary storage, and said second register continuously transfers said digitized sample transferred from said first register. after receiving and temporarily storing therein, transferring the digitized specimen immediately preceding each of said digitized specimens currently received to said third register for temporary storage therein; The three registers successively receive and temporarily store therein the digitized samples transferred from the second register, and the device further compares the values of the digitized samples each sent. a signal indicating whether one of the digitized samples has a value greater than, less than, or equal to the value of the other one sent to the comparator; comprising first and second comparator means capable of generating at its output, said first comparator means being connected to each of said first and second registers to present each register for comparison. said second comparator means is adapted to receive an indication of said digitized sample stored in said register, and said second comparator means is connected to said second and third registers to present said respective register for comparison. The signal processing means is configured to receive a representation of the stored digitized sample, and the signal processing means further determines whether the signals sent thereto are of the same sign or of different signs. ,
comprising logic means capable of generating a signal at its output indicating which condition is currently occurring, said logic means being electrically connected to said output of each of said first and second comparators; Said signal processing means receives a current signal component at its input for temporary storage and stores each successive current signal component along its storage location as each successive current signal component is received at said input. , wherein the input of the first shift register means is electrically connected to the output of the logic means, and each of the storage parts is connected to the other means. 13. The device according to claim 12, characterized in that it has: (17) storage means for said signal processing means to at least temporarily store said digitized samples provided by said analog-to-digital converter means; and said digitized samples stored in said storage means. and for the selected digitized specimens so compared, whether they are greater than, less than, or equal to the values of the other selected ones. comparator means capable of generating an indication of whether the value has a value, and a selected said indication generated by said comparator means is the same as another selected said indication given by said comparison means; and a storage means for at least temporarily storing the selected display of the logic means. 14. The device according to claim 13, characterized in that: (18) The signal processing means comprises a first, a second and a third, each of which is capable of storing the digitized sample.
registers, wherein the first register successively receives and temporarily stores therein each of the digitized samples sent by the analog-to-digital conversion means, and the first register receives and temporarily stores therein each of the digitized samples currently received therein. The digitized sample located immediately before each digitized sample is transferred to said second register for temporary storage, and said second register continuously transfers said digitized sample transferred from said first register. after receiving and temporarily storing therein, transferring the digitized specimen immediately preceding each of said digitized specimens currently received to said third register for temporary storage therein; The three registers successively receive and temporarily store the digitized samples transferred from the second register, and the device further compares the values of the digitized samples each sent. , a signal indicating whether one of the digitized samples has a value greater than, less than, or equal to the value of the other sent to the comparator. comprising first and second comparator means capable of generating an output, said first comparator means being connected to each of said first and second registers to present a respective register for comparison. said second comparator means is adapted to receive an indication of said digitized sample stored in said register, and said second comparator means is connected to each of said second and third registers to currently store said data in said respective register for comparison. The signal processing means further determines whether the signals sent thereto are of the same sign or of different signs,
comprising logic means capable of generating a signal at its output indicating which condition is currently occurring, said logic means being electrically connected to said output of each of said first and second comparators; Said signal processing means receives a current signal component at its input for temporary storage, and stores each successive current signal component along its storage location as each successive current signal component is received at said input. the first which can shift the signal components;
shift register means, wherein said input of said first shift register means is electrically connected to an output of said logic means, each of said storage sites having a storage site output connected to the other means; Claim 1 characterized by
The device according to item 4. (19) Sub-column selector means is connected to each of the storage section outputs and is capable of receiving and temporarily storing the signal components and outputting them from at least one output thereof. , and an arithmetic combination means capable of performing arithmetic operations on the signal components fed to at least one input thereof, an input of the arithmetic combination means being electrically connected to an output of the sub-column selector means. 16. The device according to claim 15, characterized in that: (20) Sub-column selector means is connected to each of the storage section outputs and is capable of receiving and temporarily storing the signal components and outputting them from at least one output thereof. , and an arithmetic combination means capable of performing arithmetic operations on the signal components fed to at least one input thereof, an input of the arithmetic combination means being electrically connected to an output of the sub-column selector means. 17. The device according to claim 16, characterized in that: